A NEW ELECTROMAGNETIC ENGINEERING PROGRAM AND ...

Progress In Electromagnetics Research B, Vol. 6, 205–224, 2008

A NEW ELECTROMAGNETIC ENGINEERINGPROGRAM AND TEACHING VIA VIRTUAL TOOLS

L. Sevgi

Electronics and Communications Engineering DepartmentDogus UniversityZeamet Sokak, No. 21, Acıbadem-Kadıkoy, Istanbul, Turkey

Abstract—The societal and technological priorities of the worldhave been continuously changing because of complex computer andtechnology-driven developments in everywhere like communications,health, defense, economy, etc. Electromagnetic (EM) systems havebecome more and more complex however the explosive growth ofcomputer capabilities has revolutionized the design and analysis ofsuch complex systems. This has made interdisciplinary exposurenecessary in modern EM engineering, also has brought up discussionsof educational challenges and novel teaching approaches that confrontwave-oriented EM engineering in the 21st century. This paper reviewsEM computer simulation strategies and summarizes novel virtual toolsthat may be used in connection with classical EM lectures as well aswith a newly proposed EM Engineering Program.

1. INTRODUCTION

Electromagnetics (EM) is everywhere! We have been witnessing thetransformation from engineering EM to EM engineering for some time.Electromagnetic engineering community must be prepared to adapt tofrequent shifts in technological priorities and rapid scientific advances,followed by rapid advances in technologies. Today, addressing thetechnical challenges posed by system complexities requires a broadrange of innovative, multi-disciplinary analytical and computationalskills that are not adequately covered in conventional EM engineeringcurricula. Universities and educational institutions have been activelyengaged in efforts to design curricula for teaching the necessary skills toa computer-weaned generation of students, with access to the internetand consequent globalization of information. Physics-based modeling,observation-based parameterization, and computer-based simulations

206 Sevgi

are the key issues of these challenges. These issues have been discussedin detail and powerful EM simulators were given in [1].

Experimentation and hands-on training are the fundamentals ofEM engineering education however strong theoretical background isalso a must. With the development of new computer technologies,interactive multimedia programming languages, and the internet itis now possible to simulate complex EM problems and laboratoryprojects of all sorts on a computer all around the world. Experiment-oriented problems can be offered without the overhead incurred whenmaintaining a full laboratory. At this point the question arises:should an intelligent balance be established between real and virtualexperimentations and how? Another similar problem is the balance tobe maintained between teaching essentials (theory) and cranking thegear (blind computer applications) [2].

Triggered by these facts, discussions and thoughts we havedeveloped and introduced multipurpose EM virtual tools [3–18] thatcan be used in most of the classical EM lectures as well as innovel EM Engineering Programs (EMEP), and have been discussingthem internationally [19–24]. We have always looked for somethingwhich would be useful, give physical insight, and effective so that theyoung EM researchers and students can use intelligently. This papertutorially reviews these simple, easy-to-use but effective virtual tools,and, in connection, proposes a novel EMEP.

2. MODELING AND SIMULATION STRATEGIES IN EM

Computational EM is a novel area that includes research in high-performance computing, applied mathematics and physics, intelligentsystems and information technologies [1]. It is a new and cost-effectiveway of solving complex EM problems beyond the reach of analyticalmethods, and the outcome is powerful software packages and virtualtools for the students, lecturers, researchers, and scientists.

Computational EM require basic understanding of fundamen-tal concepts like modeling, analytical solution, numerical solution,analytical- and numerical-based modeling, simulation, model valida-tion, code verification through canonical tests/comparisons, accredita-tion, etc. Analytical solution is the solution based on a mathematicalmodel (usually in differential and/or integral forms) of the physicalproblem in terms of known, easily computable mathematical functions,such as, sine, cosine, Bessel, Hankel functions, etc. Numerical solutionis the solution based on direct discretization of the mathematical rep-resentations by using numerical differentiation, integration, etc. Semianalytical-numerical solution is in between these two and is the solu-

Progress In Electromagnetics Research B, Vol. 6, 2008 207

tion based on partially derived mathematical forms that are computednumerically.

Modeling and simulation is extremely valuable in engineering ifbased on physics-based modeling and observable-based parameteriza-tion. It starts with the definition of the real-life problem. Its concep-tual model is the basic theory behind the real-world problem. Some dis-ciplines have already been in their mature stage in establishing theories.For example, Maxwell equations establish the mathematical model offield and circuit theories in EM engineering, define the interaction ofEM waves with matter, and form the basis for a real understanding ofelectrical problems and their solutions. All of the frequency and timedomain methods use either differential or integral form of Maxwellequations. The well-known and widely used numerical methods areFEM, MoM, SSPE, FDTD and TLM (see [1] and its references).

Finite element model (FEM) analysis of a problem includesdiscretization of the solution region into a finite number of sub-regionsor elements, derivation of the governing equations for a typical element,assembling all elements in the solution region, and solving the systemof equations obtained. A major difficulty encountered in FEM analysisis data preparation, which is essentially element generation; thereforespecial mesh generators have been developed for this purpose. Methodof Moments (MoM) is based on solving (via discretization) complexintegral equations by reducing them to a system of linear equations.The equation solved by MoM generally has the form of an electricfield integral equation (EFIE) or a magnetic field integral equation(MFIE). Although MoM can be applied also in the time domain, themajority of these equations are set in the frequency domain. The FastMultipole Method (FMM) and the Multilevel Fast Multipole method(MFMA) are the modified MoM approaches which reduce the numberof unknowns in matrix form solutions of integral representations forthe fields. The split-step parabolic equation (SSPE) technique, widelyused in propagation scenarios, implements the solution of the one-wayparabolic type wave equation. It neglects back-scatter effects and isvalid in regions close to near-axial propagation. It is an initial valueproblem, mostly used in 2D, where transverse and/or longitudinalcharacteristics can be included. The finite-difference time-domain(FDTD) method discretizes Maxwell equations by replacing derivativeswith their finite-difference approximations directly in the time domain.The transmission line matrix (TLM) method uses Huygen’s principle,i.e., it is based on circuit theory, and the discretization of fields in anarray of three-dimensional (3D) lumped elements.

208 Sevgi

3. NOVEL EM VIRTUAL TOOLS

The virtual tools we have developed and their capabilities are listedin Tables 1 and 2. These tools can be used as teaching aids in labsand projects of various EM lectures. Some virtual tools are generaland useful for major engineering disciplines. For example, D-FFT [9]is LabView-based virtual tool where the user may train themselvesvia applications of Fourier transform (FT), discrete Fourier transforms(DFT) and fast Fourier transforms (FFT). It should be rememberedthat DFT and FFT are synthetic operations so the user shouldwell-understood under what conditions DFT and FFT correspond tomathematical FT. If not satisfied, these conditions cause numericaleffects such as aliasing, spectral leakage, scalloping loss, etc., whichmay all be investigated via D-FFT virtual tool.

Table 1. Basic-level virtual EM tools developed in our group.

EM Virtual Tool Purpose D-FFT A virtual FFT instrument prepared with the LabVIEW. It can generate and

display time and frequency domain behaviors of sinusoids, a rectangular pulse, a pulse train, a Gaussian function, a sine modulated Gaussian function.

1DFDTD A Matlab-based FDTD simulation of plane wave propagation in time domain through single, double or three-layer media. EM parameters are supplied by the user.

TDRMETER A virtual time-domain reflectometer virtual tool. It is used to locate and identify faults in all types of metallic paired cable. Fourier and Laplace analyzes are also possible.

RAY / MODE HYBRID

Ray/mode representations inside a parallel plate non-penetrable waveguide. RAY serves as a tool to compute and display eigenray trajectories between specified source/observer locations and to analyze their contributions to wave fields individually. HYBRID may be used to display range and/or height variations of the wave fields comparatively, calculated via ray summation, mode field summation, and hybrid ray-mode synthesis.

DiSLAB A Matlab package designed to investigate wave propagation through a 2D dielectric waveguide. Both analytical formulations and the SSPE propagator are used for comparisons.

ARRAY A simple Matlab antenna array package of isotropic radiators accommodated with beam forming and beam steering capabilities. It plots 2D and 3D radiation patterns of a number of selected and user-located isotropic radiators.

MGL-2D A general purpose 2D FDTD package for both TE and TM type problems. Any 2D scenario may be created by the user by just using the PC mouse.

SNELL A simple Matlab package for the visualization of ray contributions between a source/receiver pair above a 2D ground using the ray shooting technique. A number of rays, whose angles of departures are specified by the user, are shot through a propagation medium characterized by various linear vertical refractivity profiles.

The packages 1DFDTD and TDRMeter simulates pulsed planewave propagation and pulsed voltage/current propagation along a


Table 2. Advanced-level virtual EM tools developed in our group.

EM Virtual Tool Purpose WEDGE A Matlab package for the exploration of wave propagation inside a 2D non-

penetrable, homogeneously filled wedge-waveguide. It is designed to investigate line-source-excited wave fields in terms of normal, adiabatic and intrinsic mode solutions.

DRMIX A Matlab-based Millington package prepared for the mixed-path path loss predictions. The effects of the number of multi-mixed paths, path-lengths, electrical parameters of each propagation section, and the frequency can be investigated.

GrSSPE A simple Matlab groundwave propagation package for the visualization of EM propagation over non-flat terrain through non-homogeneous atmosphere, for waves radiated by a horizontally oriented antenna over the ground.

GrMoMPE A Matlab package which modifies MoM method by the application of forward backward spectral acceleration (FBSA) technique and integrate it with the SSPE method. Precise SSPE vs. MoM comparisons are possible.

MSTRIP An FDTD-based EM simulator for the broadband investigation of microstrip circuits. The user only needs to picture the microstrip circuit via computer mouse on a rectangular grid, to specify basic dimensions and operational needs such as the frequency band, simulation length.

MGL-RCS An FDTD-based EM simulator for RCS prediction. The user only needs to locate a 3D image file of the target in 3DS graphics format, specify dimensions and supply other user parameters. The simulator predicts RCS vs angle and/or RCS vs. frequency.

transmission line, respectively. These tools are important for thevisualization of 1D pulse scattering. They can also be used to show theanalogy between plane waves and transmission lines. They both arebased on FDTD modeling of first order coupled differential equationsof the same form. For the plane waves, the coupled equations relatespatial variations of the electric fields to the time variations of themagnetic fields (or vice versa) in terms of medium EM parametersconductivity, permittivity and permeability. The same equations relatepulsed voltage variations along the transmission line to the pulsedcurrent variations in time via the line parameters modeled by thelumped elements RLC. These virtual tools are also very effectivein showing the time and the frequency domain characteristics ofplane waves and transmission lines using DFT or FFT. The Laplacetransform is also used in the TDRMeter to show the transient effectsalong the line [12]. Figs. 1 to 4 show some examples obtained usingthe TDRMeter tool.

As shown in Fig. 3(a), the standard Fourier Transformationprocedure includes (i) discrimination of the incident and reflected pulsein the time domain, (ii) moving to the frequency domain via FFT, and(iii) taking the ratio of the reflected and incident pulses within theband of the pulse. Similarly, the shape of the echo may be used to

210 Sevgi

Figure 1. The front panel of the TDRMeter package, resistivetermination, a Gaussian pulse traveling towards the load (a 50 Ωtransmission line, Rs = 10 Ω, RL = 100 Ω).

1

4

2 3

5 6

Figure 2. Gaussian pulse injection along a transmission line short-circuited at both ends. Graphs show different time instants (1: pulseinjection, 2: detachment of the right- and left-traveling pulses, 3,4: pulses propagating towards left and right, 5, 6: right-propagatingpulses, one from the source, the other reflected from the left boundarywith the opposite phase).

predict the nature of the mismatch along the transmission line and/orat the load. This is achieved (as in Fig. 4) by the application of LaplaceTransformation procedure [12].


nHL

pFC

R

L

L

L

10

5

50

500Zincident reflected

(a)

(b)

Ω

Ω=

=

=

=

Figure 3. (a) Signal vs. time at mid point of a 50 Ω, 0.5 m TLunder parallel RLC termination (RL = 50 Ω, CL = 5 pF, LL =10 nH). Matched termination is used at the source (Rs = 50 Ω).Pulse length = 400 ps, (b) Voltage reflection coefficient vs. frequencyobtained with the FFT procedure; Solid: TD simulation result,Dashed: Analytical exact solution.

The TDRMeter virtual tool can be used to teach, understand,test and visualize the TD TL characteristics; echoes from variousdiscontinuities and terminations. Inversely, the types and locationsof fault can be predicted from the recorded data. Any electronicdevice and/or circuit include TL lines (i.e., coaxial, two-wire cables,parallel plate lines, microstrip lines, etc. therefore have transient EMcharacteristics. A wide range of EMC problems may be simulated withthe TDRMeter virtual tool, such as, conducted emissions, common anddifferential mode effects, etc., both in time and frequency domains.

212 Sevgi

nHL

R

L

L

27

50

500Z

nHLL 2.26Plot:

t=0t=

1.368

τ

=

=

=

= Ω

Figure 4. Signal vs. time (step response) at mid point of a 50 Ω,0.5 m TL under serial RL termination (RL = 50 Ω, LL = 27 nH).Matched termination is used at the source (Rs = 50 Ω). The valueof the inductor calculated from the plot is 26.2 nH.

Also, reflections due to various types of transmission structures,impedance effects of vias, signal and power line couplings, etc., onprinted circuits with smaller and smaller dimensions make signalintegrity one of the most important and complex EMC problems.Therefore, the virtual tool may be particularly helpful for a student orcircuit designer in the evaluation of signal integrity.

ANTEN [14] is a Matlab package to visualize beam forming andbeam steering capabilities of various linear, circular, and planar arraysof isotropic radiators. The user may form any kind of an array andobserve radiation patterns in 2D or 3D (for vertical or horizontalpolarizations). An interesting feature is added in designing planararrays. The user may design an arbitrary array by just locatingarbitrary number of isotropic radiators using the PC mouse. Inter-element phasing is done automatically to steer the beam of the designedarray to the specified beam pointing direction. An example producedvia ANTEN is given in Fig. 5.

The packages RAY/HYBRID [15], WEDGE [16] and DiSLABare designed in Matlab using mathematical solutions, therefore arevery important to teach ray-mode representations in 2D parallel plate,wedge-type and slab waveguides, respectively. TM characteristics ofthese three structures are almost canonical and covered in most of theclassical EM lectures. To teach and show mode solution, ray solutionsand their hybrid usage inside a 2D parallel plate waveguide RAY and


(a) (b)

Figure 5. A 10×10 planar array, (a) horizontal plane 2D pattern,(b) 3D radiation pattern (f = 300 MHz, inter-element distances are0.25 m in both x- and y-directions, the beam angles are θ0 = 45 andϕ0 = 45.

HYBRID virtual tools are very useful. Two examples are presentedin Figs. 6 and 7. As shown, the user may visualize individual orcollective contributions of the rays and/or the modes. The same istrue for the WEDGE and DiSLAB virtual tools. The 2D parallelplate waveguide with penetrable and non-penetrable boundariesare canonical structures in teaching rays representation, eigenvalue(source-free) and Green’s function (source-driven) problems [15, 16].

SNELL [17], MGL-2D [11], GrSSPE [13], GrMoMPE [5] andDrMIX [6] are 2D propagators based on different frequency or timedomain methods. One- and/or two-way propagation problems, eitherin the time or in the frequency domains, can be simulated. SNELLis a ray shooting virtual tool which uses only a simple equation. Theuser specifies refractivity variations of the medium, may locate non-penetrable obstacles on the ground along the propagation direction,vertical beam of the source. The 2D propagation environment isdivided into a number of horizontal slices and a number of rays areshot and Snell law is applied at each range/height point. An exampleis given in Fig. 8.

214 Sevgi

j=1

j=2

j=3

j=4

Figure 6. (Left) first four eigenarray inside a 1 m-wide parallelplate waveguide for a set of parameter (distance= 5.6 m, sourceheight = 0.3 m, observer height = 0.7 m, f = 2387 MHz, ka = 50,n = 1). The number of propagating modes for these parameters is15. (Right) Wave field vs. height calculated via both mode and raysummations (mode summation is the reference solution).

Figure 7. Same with the Fig. 6 but for n = 1 to 5. As observed, raysolution approaches to the reference (mode) solution as the rays stuffthe waveguide.

GrSSPE and GrMoMPE are MoM- and SSPE-based propagatorswhere the user may design arbitrary non-flat terrain profiles andcalculates signal strength and path loss variations along the range upto a specified receiver point. The user may also import longitudinal


Figure 8. The front panel of the Matlab package SNELL andray shooting through an environment with a vertical tri-linearrefractivity profile having two obstacles along the propagation path(source height = 50 m, maximum height = 300 m, layer height = 0.2 m,maximum range = 2500 m, first ray angle = 45, last ray angle = 120,ray increment = 0.375, Refractivity slope = 0.001, the heights ofducting and anti-ducting are 80 m and 150 m, respectively, 1.obstacle: height = 50 m, base length = 200 m, range = 1000 m; 2.obstacle: height = 100 m, base length = 500 m, range = 1500 m).

terrain profiles and run the simulations over these profiles and obtainpath loss vs. range curves. This is really important, since itgives a possibility of comparing a real-life measurement data againstsimulation data obtained via the SSPE and/or MoM propagators.Finally, MGL-2D is an FDTD based 2D propagator which can beused for broad range of EM problem simulations. Some examples arepresented in Figs. 9–11.

Finally, MSTRIP [10] and MGL-RCS [4] are 3D FDTD codesdesigned specifically to investigate single- or double-layer microstripcircuits and EM reflectivity of objects, respectively. MSTRIP is aFDTD-based EM simulator accommodated with the powerful perfectlymatched layer (PML) routine, specifically designed for microstripcircuit investigations. The user only needs to picture the microstripcircuit via computer mouse on a rectangular grid, to specify basicdimensions and operational needs such as the frequency band,

216 Sevgi

Figure 9. The SSPE virtual tool and propagation through tri-linear atmosphere over an irregular terrain; frequency 100 MHz,maximum range 50 km, maximum height 1500 m, transmitter height250 m, antenna tilt 2 degrees upwards, and beamwidth 1 degrees.Linearly increasing and decreasing refractivity segments have slopesof 5000 M/km.

Figure 10. The MGL-2D virtual tool and outdoor propagationmodeling along a horizontally-projected user-specified street after sometime.


Figure 11. A typical non-flat terrain scenario (a surface with manysmooth irregularities; with range in [km] and height in [m]) andpropagation factor in dB vs. range in [km] for the TM-type problem(f = 15 MHz, maximum range = 20 km, transmitter height = 400 m,receiver height = 10 m).

Figure 12. The MSTRIP virtual tool; a typical 4-port coupler, asnapshot during its FDTD simulation and the frequency responses atfour ports.

218 Sevgi

Figure 13. The front panel of the MGL-RCS package and the 3Ddiscrete model of an air target (a 50 m-long DC-10 jet) imported froma 3DS file.

simulation length. The rest is handled by the virtual tool and outputsuch as signal vs. time signal vs. frequency, etc., can be displayed on-line during the simulation. Moreover, three dimensional visualizationsare possible where the user may record any length of films for furtherillustrations. MGL-RCS is an RCS prediction tool which can be usedto create 3D targets from canonical blocks such as rectangular prism,cone, cylinder, sphere, and torus; or 3D models of surface and/or airvehicles saved in different graphics files (e.g., 3DS format) can be usedas radar targets. Near EM scattered fields around the target, causedby an incident Gaussian plane wave hitting the target from a specifieddirection, can be simulated using a 3D numerical FDTD model. EMtransient scattering may also be recorded in a specified file as a videoclip. Far fields can be extrapolated and stored via the application ofequivalence principle and bi-static RCS patterns can be plotted usingdiscrete Fourier transformation (DFT) onto the stored time-domainEM scattered data. Figs. 12 to 14 are reserved for the examples ofMSTRIP and MGL-RCS.


f=80MHz

18 dB

=4f=40MHz

30 dB

=2 λ λ

(a) (b)

Figure 14. FDTD vs. NEC for angular bi-static RCS predictionsat two different cases; (a) f = 40 MHz, θi = 90, ϕi = 0, ϕs = 0,0 ≤ θs ≤ 360 with angular resolution of ∆θ = 3, Vertical scan, σϕϕ-case. (b) f = 80 MHz, θi = 90, ϕi = 90, ϕs = 90, 0 ≤ θs ≤ 360with angular resolution of ∆θ = 3, Vertical scan, σϕϕ-case. Solid(red): FDTD, Dashed (blue): NEC2. Maximum plot values are givenon top right.

4. A NOVEL EM ENGINEERING PROGRAM — EMEP

Most of ECE engineering program have EM lectures such as EM Fields,EM Waves, Antennas and Propagation, Introduction to MicrostripCircuits, Wireless Communication, EMC Engineering, Filter Theory,Radar and Sensing Systems, etc. However, current complexity of mostof the EM problems necessitates revisit of the number and contentsof these lectures. A novel EMEP program, partially based on thepresented virtual tools can be designed with a reduced number oflectures and credits. Essential topics can be covered in four differentlectures; EM Engineering I, II, and III, Microwave Engineering. Otherrelated engineering topics may be covered in the lecture EM SpecialTopics. Fundamentals topics to be covered inside the first four lecturesare listed in Tables 3 and 4.

Junior-level EM courses basically covers EM topics in one-and two-dimension. After summarizing current challenging EMproblems, and reviewing Fourier and discrete Fourier fundamentals,the first quarter (nearly three weeks) of the first lecture EM Eng-Iis reserved for Maxwell equations, fundamental representations andbasic derivations. Then, EM fields/waves are discussed in 1D in the

220 Sevgi

Table 3. Junior-level EM lectures and topics to be covered in theproposed EMEP.

EM Engineering - I EM Engineering & Challenging ProblemsFourier transforms, DFT & FFT LabView-based FFT VTMaxwell EquationsTime & Frequency RepresentationsElectrostatics and Magnetostatics Electrodynamics Scattering, Radiation & Diffraction

Spherical waves, Cylindrical wavesPoynting theorem & EM Power Wave/matter interaction Plane Waves & 1D FDTD SolutionsMatlab-based 1D FDTD VTPlane waves Tran smission Line AnalogyTransmission Line TheoryMatlab-based TDRMeter VT

EM Engineering - II Complex 2D EM Problems 2D Parallel Plate Waveguide & Ray/ModeSolutions (HYBRID VT) 1D Sturm-Liouville Prolem (STURM VT) Ray tracing, Ray Shooting & SNELL VTFriis Formulae & Path Loss Definiton Propagation Through Atmosphere,Refractivity Effects & SSPE VTDielectric Slab Waveguide & DiSLAB

FDTD-based MGL-2D VTEM (Hertz & ring) Dipoles Antenna Fundamentals EMC & Communication AntennasArrays, Beam Forming& SteeringMatlab-based ARRAY VTWire Antennas & NEC Modeling3D Rectangular waveguidesResonators

........

.........

.........

_

........

Table 4. Senior-level EM lectures and topics to be covered in theproposed EMEP.

EM Enginering - III Fundamentals of EMC, EMI & BEMSignals and Noise in Electromagnetics Non-Linearity Effects & Harmonics EMC Standards, Tests & Measurements EMC & Shielding Effectiveness (SE VT) Specific Absorption Rate (SAR VT)

N-port devices & S-parameters LC TL Equivalence Microstrip Lines & Microstrip Circuits Broadband Filter Design Chebychev and Butterworth Filters FDTD-based 3D MSTRIP VT

EM Engineering - IV (Special Topics)Computational Electromagnetics Finite Difference Time Domain Method Finite Element Method Method of Moments Radars and Sensing Systems

Detection & Tracking Algorithms RCS Modeling & Simulation MGL-RCS VT Multi-Sensor Surveillance systems Opical Fibers, Wireless Communication

......

......

.....

.....

_


second quarter; which focuses on the plane waves. The transmissionline theory is summarized in the third quarter starting with thediscussion of plane wave — transmission line analogy. The lastquarter is reserved for the virtual lab experimentations using thetwo virtual tools, 1DFDTD and TDRMeter [12]. Resonance effectsof the plane wave propagation in a 1D finite region and/or along afinite length short/open circuited transmission line can be presentedusing the discrete Fourier transformation tools DFT or FFT. Also,transients along a transmission line can be presented using the Laplacetransformation and the TDRMeter data. Numerical experimentationsof the excitation and/or observation of the resonances (i.e., modes) in1D can be performed easily.

Fundamental EM problems in 2D are included in the secondlecture, EM Eng–II. Classical 2D parallel plate waveguide withpenetrable as well as non-penetrable boundaries are discussed inthis course. Ray, mode and hybrid solutions are presented togetherwith the visual capabilities of the RAY, HYBRID [15] and DiSLABvirtual tools. Ray-mode representations and ray shooting methods arecovered in this lecture. The relation between eigenvalue and Green’sfunction problems are related via the Sturm-Liouville problem. TheFDTD-based virtual tool MGL-2D [11] can be used to create manyrealistic EM scenarios, from indoor-outdoor propagation to tunneling,waveguides to resonators, etc. Antennas and radiation topics are alsocovered in this lecture.

The first advanced level EMEP lecture EM Eng-III is basicallyreserved for EM compatibility, bio-EM engineering, and microstripcircuit design and analysis. Signal processing aspects in EM arealso covered in this lecture. Finally, the last lecture Special EMTopics is reserved for modeling and numerical simulations, and mayinclude radar and sensing systems, problems related to Multi SensorSurveillance, fundamentals of Wireless Systems, and Fiber Optics. Alllectures are 3-credit (2 hr course +2 hr applications with virtual tools).The total number of credits of this EMEP with these 4 lectures is 12and replaces those classical EM lectures mentioned above with the totalnumber of credits of more than 21. This gives students a plenty of spaceto work with engineering projects. It should be noted that, the newEMEP necessitates revision of two more junior-level courses; NumericalAnalysis and Probability, Statistics, and Stochastic Modeling.

222 Sevgi

5. CONCLUSIONS

A new EM Engineering Program is introduced to cope with currentchallenges in engineering education. The program is based on twojunior and two-senior level EM Engineering courses. Various powerfulvirtual tools which play essential roles in teaching in this new EMEPare also discussed. It should be noted that, an intelligence balancebetween EM theory and numerical simulations must be establishedwhen using this EMEP.

REFERENCES

1. Sevgi, L., Complex Electromagnetic Problems and NumericalSimulation Approaches, IEEE Press & John Wiley and Sons Inc.,NJ, June 2003.

2. Sevgi, L. and I. C. Goknar, “An intelligent balance in engineeringeducation,” IEEE Potentials Magazine, Vol. 23, No. 4, 40–41,Oct./Nov. 2004.

3. Uluısık, C., G. Cakır, M. Cakır, and L. Sevgi, “RadarCross Section (RCS) modeling and simulation: Part I —Definitions, strategies, and canonical examples,” IEEE Antennasand Propagation Magazine, Vol. 50, No. 1, Feb. 2008.

4. Cakır, G., M. Cakır, and L. Sevgi, “Radar Cross Section (RCS)modeling and simulation: Part II — A novel FDTD-basedRCS prediction virtual tool,” IEEE Antennas and PropagationMagazine, Vol. 50, No. 2, Apr. 2008.

5. Akleman, F. and L. Sevgi, “A novel MoM- and SSPE-basedgroundwave propagation field strength prediction simulator,”IEEE Antennas and Propagation Magazine, Vol. 49, No. 5, 69–82, Oct. 2007.

6. Sevgi, L., “A mixed-path groundwave field strength predictionvirtual tool for digital radio broadcast systems in medium andshort wave bands,” IEEE Antennas and Propagation Magazine,Vol. 48, No. 4, 19–27, Aug. 2006.

7. Gunduz, S., G. Cakır, and L. Sevgi, “A generic microstripstructure for the realization of all-type broadband filters,”Microwave and Optical Technology Letters, Vol. 48, No. 12, 2390–2393, Dec. 2006.

8. Sevgi, L., “Virtual tools/labs in electrical engineering education,”ELEKTRIK, Turkish J. of Electrical Engineering and ComputerSciences, Vol. 14, No. 1, 113–127, Special issue on Electrical and


Computer Engineering Education in the 21st Century: Issues,Perspectives and Challenges, 2006.

9. Sevgi, L. and C. Uluısık, “A labview-based virtual instrumentfor engineering education: A numerical fourier transform tool,”ELEKTRIK, Turkish J. of Electrical Engineering and ComputerSciences, Vol. 14, No. 1, 129–152, Special issue on Electrical andComputer Engineering Education in the 21st Century: Issues,Perspectives and Challenges, 2006.

10. Cakır, G., M. Cakır, and L. Sevgi, “A novel virtual FDTD-basedmicrostrip circuit design and analysis tool,” IEEE Antennas andPropagation Magazine, Vol. 48, No. 6, 161–173, Dec. 2006.

11. Cakır, G., M. Cakır, and L. Sevgi, “A multipurpose FDTD-basedtwo dimensional electromagnetic virtual tool,” IEEE Antennasand Propagation Magazine, Vol. 48, No. 4, 142–151, Aug. 2006.

12. Sevgi, L. and C. Uluısık, “A matlab-based transmission linevirtual tool: Finite-difference time-domain reflectometer,” IEEEAntennas and Propagation Magazine, Vol. 48, No. 1, 141–145,Feb. 2006.

13. Sevgi, L., C. Uluısık, and F. Akleman, “A matlab-based two-dimensional parabolic equation radiowave propagation package,”IEEE Antennas and Propagation Magazine, Vol. 47, No. 4, 184–195, Aug. 2005.

14. Sevgi, L. and C. Uluısık, “A matlab-based visualization packagefor planar arrays of isotropic radiators,” IEEE Antennas andPropagation Magazine, Vol. 47, No. 1, 156–163, Feb. 2005.

15. Felsen, L. B., F. Akleman, and L. Sevgi, “Wave propagation insidea two-dimensional perfectly conducting parallel plate waveguide:Hybrid ray-mode techniques and their visualisations,” IEEEAntennas and Propagation Magazine, Vol. 46, No. 6, 69–89,Dec. 2004.

16. Sevgi, L., F. Akleman, and L. B. Felsen, “Visualizations of wavedynamics in a wedge-waveguide with non-penetrable boundaries:Normal, adiabatic, and intrinsic mode representations,” IEEEAntennas and Propagation Magazine, Vol. 49, No. 3, 76–94,Jun. 2007.

17. Sevgi, L., “A ray shooting visualization matlab package for2D ground wave propagation simulations,” IEEE Antennas andPropagation Magazine, Vol. 46, No. 4, 140–145, Aug. 2004.

18. Sevgi, L., “Modeling and simulation strategies for electromagneticwave propagation in complex environments: Groundwave pathloss prediction virtual tools,” IEEE Transactions on Antennasand Propagation, Vol. 55, No. 6, 1591–1598, Special issue on

224 Sevgi

Electromagnetic Wave Propagation in Complex Environments: ATribute to Leopold Benno Felsen, Jun. 2007.

19. Uluısık, C. and L. Sevgi, “Numerical modeling and simulationstudies of 2D radiowave propagation over non-flat terrain andthrough inhomogeneous atmosphere,” Springer Proceedings inPhysics Series, Vol. 104, 45–54, Complex Computing Networks,Jan. 2006.

20. Sevgi, L., “From engineering electromagnetics towards electro-magnetic engineering,” ELECO 2007, 5th International Confer-ence on Electrical and Electronics Engineering, Bursa, Turkey,Dec. 5–9, 2007.

21. Sevgi, L., “Electromagnetic engineering and teaching via virtualcomputer tools,” EUROCON 2007 the International Conferenceon Computer as a Tool, a half-day short course, Warsaw, Poland,Sept. 9–12, 2007.

22. Sevgi, L., “Modeling and simulation strategies in antennas andpropagation: Novel virtual tools,” EuCAP 2006, the First ESAEuropean Conference on Antennas and Propagation, a half-dayshort course, Nice-France, Nov. 6–10, 2006.

23. Sevgi, L., “Modeling and simulation strategies in electromagnet-ics: Teaching via virtual tools,” IEEE AP-S International Sympo-sium & USNC/URSI National Radio Science AMEREM Meeting,a half-day short course, Albuquerque, New Mexico, USA, Jul. 9–14, 2006.

24. Sevgi, L., “Modeling and simulation strategies in EMC engineer-ing: Problems and challenges,” EMC 2005 6th International Sym-posium on EMC and EM Ecology, two-hour tutorial, St Peters-burg, Russia, Jun. 21–24, 2005.

A NEW ELECTROMAGNETIC ENGINEERING PROGRAM AND ...

Documents

Transcript of A NEW ELECTROMAGNETIC ENGINEERING PROGRAM AND ...